Method of Determining Database Search Path

1. A method of determining a database search path in a database system, which includes databases each containing records with search keys of related records stored in other databases, for extracting records containing a search key that is entered into the system, in which system a starting record extraction is carried out using a starting search key that is entered into a starting database thereby extracting records from the starting database, and an intervening search key that is contained in the extracted records from the starting database and that is different from the starting search key is entered into the intervening database different from the starting database so as to carry out an intervening record extraction in the intervening database in a chain-reactive manner form the starting database via a plurality of intervening databases to the terminal database using search keys and records to link the starting, intervening, and terminal databases, the extracted records as search results, the method comprising: a first step of classifying the plurality of databases into groups based on the characteristics of data included in individual databases; a second step of creating a correlation network wherein the groups are related with each other based on relevance between the groups; a third step of obtaining inter-group path candidates in the correlation network; and a fourth step of designating a starting database and a terminal database from the plurality of databases and obtaining an inter-database path candidate that exists on one of the inter-group path candidates that exists between the starting and terminal databases.

2. The method according to claim 1 , wherein the third step includes giving priority to the inter-group path candidates and the fourth step includes following a higher priority inter-group path candidate to derive said inter-database path candidate that exists on the higher priority inter-group path candidate.

3. The method according to claim 2 , wherein higher priority is given to links between the groups that constitute a path that has to be passed through based on a pre-determined characteristic correlation between databases, and normal priority is given to remaining links.

4. A method of determining a database search path in a database system, which includes databases each containing records with search keys of related records stored in other databases, for extracting records containing a search key that is entered into the system, in which system a starting record extraction is carried out using a starting search key that is entered into a starting database thereby extracting records from the starting database, and an intervening search key that is contained in the extracted records from the starting database and that is different from the starting search key is entered into the intervening database different from the starting database so as to carry out an intervening record extraction in the intervening database in a chain-reactive manner form the starting database via a plurality of intervening databases to the terminal database using search keys and records to link the starting, intervening, and terminal databases, the extracted records as search results, the method comprising: a first step of classifying the plurality of databases into groups based on the characteristics of data included in individual databases; a second step of creating a correlation network wherein the groups are related with each other based on relevance between the groups; a third step of designating a starting database and a terminal database from the plurality of databases and obtaining inter-group path candidates that exists between the starting and terminal databases in the correlation network; and a fourth step of obtaining an inter-database path candidate that exists on one of the inter-group path candidates.

5. The method according to claim 4 , wherein the third step includes giving priority to the inter-group path candidates and the fourth step includes following a higher priority inter-group path candidate to derive said inter-database path candidate that exists on the higher priority inter-group path candidate.

6. The method according to claim 5 , wherein higher priority is given to links between the groups that constitute a path that has to be passed through based on a pre-determined characteristic correlation between databases, and normal priority is given to remaining links.